This project was aimed at classifying motor imagery from EEG data using machine learning to play a game like pong. To do so, dataset IV from the 4th BCI competition was used. It is an asynchronous controller and outputs a motor imagery scale continuously. This allows the user to perform an action at will and is not time-locked.

An example output is shown below. Closer to 1 indicated right foot imagery, closer to -1 indicates closer to left hand imagery.